READMem: Robust Embedding Association for a Diverse Memory in Unconstrained Video Object Segmentation

TL;DR

READMem introduces a memory management framework for semi-automatic video object segmentation that dynamically maintains diverse and non-redundant memory content, improving efficiency and performance on long videos.

Contribution

It proposes a novel memory update strategy based on embedding diversity and robust association, reducing redundant data and memory size in long-term video segmentation.

Findings

Achieves competitive results on the Long-time Video dataset (LV1).

Maintains performance on short sequences while optimizing long video processing.

Reduces memory demands without sacrificing segmentation accuracy.

Abstract

We present READMem (Robust Embedding Association for a Diverse Memory), a modular framework for semi-automatic video object segmentation (sVOS) methods designed to handle unconstrained videos. Contemporary sVOS works typically aggregate video frames in an ever-expanding memory, demanding high hardware resources for long-term applications. To mitigate memory requirements and prevent near object duplicates (caused by information of adjacent frames), previous methods introduce a hyper-parameter that controls the frequency of frames eligible to be stored. This parameter has to be adjusted according to concrete video properties (such as rapidity of appearance changes and video length) and does not generalize well. Instead, we integrate the embedding of a new frame into the memory only if it increases the diversity of the memory content. Furthermore, we propose a robust association of the embeddings stored in the memory with query embeddings during the update process. Our approach avoids the accumulation of redundant data, allowing us in return, to restrict the memory size and prevent extreme memory demands in long videos. We extend popular sVOS baselines with READMem, which previously showed limited performance on long videos. Our approach achieves competitive results on the Long-time Video dataset (LV1) while not hindering performance on short sequences. Our code is publicly available.